Predisposition Testing for Inherited Breast Cancer

Predisposition Testing for Inherited Breast Cancer

ABSTRACT: Predisposition testing (ie, genetic testing that provides information about a person’s susceptibility to disease) is now available for several inherited forms of cancer. Individuals who are found to have an altered gene (eg, a germ-line mutation in a cancer susceptibility gene) have a higher risk of developing cancer than those who do not carry an altered gene. Therefore, predisposition testing can be a powerful clinical tool for assessing a person’s risk for developing cancer. All health care providers, particularly cancer care providers, should be knowledgeable about cancer predisposition testing options. This article provides an overview of predisposition testing for inherited breast cancer, including general facts about testing, potential risks and benefits, specific genetic counseling issues, and molecular details of known breast cancer susceptibility genes. [ONCOLOGY 12(8):1227-1241, 1998]


Breast cancer is a major health problem in the United States and in
other industrialized nations. In 1997, an estimated 180,000 women
were diagnosed with breast cancer, and an estimated 44,000 women died
from the disease.[1] Current predictions hold that 1 of 8 women who
live to 85 years of age will develop breast cancer, compared with 1
of 11 in 1980, and 1 of 16 in 1970.

Early detection of breast cancer rose significantly during the first
half of the 1980s because of the introduction of screening
mammography. Diagnosis of regional disease in the breast and adjacent
tissues, which is 72% curable, increased by 11% in women under 50
years of age and by 6% in those over age 50. The detection of
localized cancer, which is 93% curable, rose by 11% and 39%,
respectively, in these two age groups. However, despite the marked
increase in the total number of diagnosed cases, breast cancer
mortality has remained relatively stable and has even declined in
some areas of the country.[1,2]

A promising approach for further reducing the high incidence and
mortality associated with breast cancer lies in better understanding
the risk factors for the disease and implementing strategies to
reduce those risk factors. Breast cancer likely develops as a result
of interactions between the environment and the genes that cause
increased cancer susceptibility.[3] It is currently estimated that 5%
to 10% of all breast cancer cases are attributable to the inheritance
of highly penetrant mutations in breast cancer susceptibility genes.[4]

Two such genes, BRCA1 and BRCA2, have been well characterized. Over
the next few years, it is likely that predictive testing for
hereditary breast and ovarian cancer will become part of the medical
management of women at high risk. Longitudinal follow-up studies of
women who are genetically predisposed may provide insights into the
underlying pathogenesis of all forms of breast cancer, which should
eventually lead to improved diagnosis and treatment of the disease.

Breast Cancer Risk Assessment

Several risk factors are known to contribute to the development of
breast cancer. Table 1 outlines
some of these factors. Clearly, the greatest risk factors for breast
cancer are increasing age and a positive personal or family history
of the disease. For example, the age-related risk for developing
breast cancer is 1 in 2,500 (0.04%) for a 30-year-old woman, 1 in 50
(20%) for a 50-year-old, and 1 in 11 (9%) for a 75-year-old.

Although an estimated three out of four breast malignancies occur in
women over 50 years old, a significant percentage occur at young ages
and may be due to a genetic predisposition.[2] Health care providers
should know how to recognize young individuals at risk who are
unlikely to receive breast cancer screening as part of their routine
medical care. They should also know when genetic testing may be
useful in quantifying breast cancer risk and when patients should
simply be reassured.

Empiric Risk Models

A woman’s cumulative risk of breast cancer can be estimated from
data derived from different epidemiologic studies.[5,6] Tabular risk
data compiled by Claus et al using information from the Cancer and
Steroid Hormone (CASH) study can be readily applied to clinical
situations.[5] A particular advantage of the CASH model is that it
factors in age of onset of affected relatives, which has been shown
to have a strong effect on risk.

The second tool available for counseling, the Gail model, is based on
information gathered from the Breast Cancer Detection and
Demonstration Project.[6] The Gail model uses five variables to
calculate risk ratios: current age, age at first live birth, age at
menarche, number of first-degree relatives with breast cancer, and
number of breast biopsies. Tables and computer programs are available
to estimate individualized age-specific risk based on this model.

The Gail model does not provide an accurate means of risk assessment
in all cases because it does not consider the presence of breast
cancer in second-degree relatives, nor does it take into account
cases of ovarian cancer in the family. The Claus model appears to be
more closely applicable to women with inherited breast cancer because
it incorporates age of disease onset in affected relatives. Even
though both models underestimate the risk for individuals from
families with inherited breast cancer who are gene carriers, while
overestimating the risk in noncarriers, they are clinically useful
tools. The Gail model was used to assess risk in the tamoxifen
(Nolvadex) breast cancer prevention trial (Case

Breast Cancer Susceptibility Genes

Genetic predisposition to breast cancer is likely due to a
constellation of polygenic, multifactorial, and single-gene
disorders. At least four genes, BRCA1, BRCA2, TP53, and PTEN, have
been identified and shown to be mutated in families with a genetic
predisposition to breast cancer (Table 2)
Several features of inherited breast cancer can aid the clinician in
identifying individuals who may have a genetic predisposition for the
disease. These include: early age of onset (< 45 years old),
bilaterality, multiple affected members in different generations in
the family, and an association with other cancers, such as ovarian
cancer and sarcoma.


The initial investigations in families with a high incidence of
breast cancer only or breast and ovarian cancers led to the discovery
of a single autosomal-dominant cancer susceptibility gene, BRCA1.
This gene, localized on chromosome 17q12-21.1, is thought to account
for 30% to 45% of breast cancer cases in families with a high
incidence of early-onset breast cancer and nearly 90% of cases in
families with a high incidence of breast and ovarian cancers.[7,8]
The BRCA1 gene was identified by positional cloning methods in 1994,
and deleterious predisposing mutations have been detected in many
kindreds with inherited breast cancer.[9,10]

The BRCA1 gene spans a genomic region of almost 100 kilobase (kb) in
length and contains 24 exons. The full-length messenger RNA (mRNA) is
7.8 kb, encoding a protein of 1,863 amino acids. More than 340
mutations and sequence variations have been detected, and it is quite
clear that not all mutations have yet been discovered.


In December 1995, the second cancer susceptibility gene, BRCA2, was
isolated on chromosome 13q12-13. Using families with multiple cases
of early-onset breast cancer showing evidence against linkage to
BRCA1, Wooster et al performed linkage analysis to isolate the BRCA2
gene.[11] This gene appears to account for about 35% of families with
early-onset breast cancer.[11-13] Several other cancers appear to be
part of the BRCA2 spectrum, including pancreatic, fallopian tube,
laryngeal, uterine, and male breast cancers, as well as adult leukemia.[14,15]

The BRCA2 gene is composed of 27 exons distributed over roughly 70 kb
of genomic DNA, encoding a protein of 3,418 amino acids. Although
BRCA2 shows no homology to BRCA1, both genes have a large exon 11,
translational start sites in exon 2, and coding sequences that are
rich in A-T (adenine and thymine) sequences. Also, both BRCA1 and
BRCA2 span approximately 70 kb of genomic DNA and are expressed at
high levels in the testis.

The fact that BRCA2 has a different spectrum of associated cancers,
including male breast cancer, suggests that the two genes may not
function in the same genetic pathway. The mutational profile of BRCA2
differs from that of BRCA1 as well (see Figure
). Microinsertions and point mutations are equally common in
the BRCA1 gene, whereas microdeletions predominate in BRCA2.[13]

BRCA Mutational Spectrum

Different types of DNA alterations lead to deleterious mutations in
the BRCA genes. A change in a single amino acid that does not affect
the manner in which the remainder of the protein is translated is
called a missense mutation. The degree that missense mutations
contribute to cancer development is unpredictable because some
missense mutations do not alter the physical properties of the
protein. When a base substitution alters an amino acid, resulting in
the production of a stop codon (TAA, TAG, or TGA), a nonsense mutation
has occurred, causing protein translation to terminate at that point.

Frameshift mutations occur when either one or a few
nucleotides are inserted or deleted in the coding region of the gene.
These mutations alter the triplet code for all of the codons that
follow and, thus, completely change the sequence of amino acids. A
truncated protein caused by a nonsense or frameshift mutation will
usually result in a defective protein product.

Variations can also occur in the gene’s noncoding region,
leading to a reduction or loss of protein synthesis from the mutant
chromosome. These noncoding mutations usually occur outside of
the coding sequence of a gene and can be due to either
microdeletions, large deletions, nucleotide insertions/deletions, or
substitutions. A few noncoding mutations have been described in the
BRCA genes.

The final category of mutation, intron/exon splice site mutations,
can result from either single-base changes or the insertion or
deletion of one or more nucleotides in the intron/exon boundary.
Splicing mutations may cause the production of a nonfunctioning
protein; however, there are some splice site variants that may not
alter protein function.

Missense, nonsense, and frameshift mutations represent more than 80%
of the total mutations described thus far in the BRCA1 gene.[11]

One striking feature of the spectrum of BRCA1 mutations described to
date is that most (55%) appear in exon 11, which is also the largest
exon. Other common recurring mutations are found in exons 2 (5.5%), 5
(4.7%), and 16 (4.7%).[16] If exon 11 is divided into three equal
parts, an increased number of different mutations can be found toward
the 3´ end of the exon.

A mutation database was established in 1995 by the National
Institutes of Health (NIH) on the World Wide Web to serve as a
repository for researchers with information on all of the known
mutations and primers for amplifying the BRCA genes (see Figure
and Figure 2). This Web site
also has a bulletin board for discussions.[17]

Studies in Ashkenazi Jewish Individuals--The recognition of an
unexpectedly high frequency of a specific mutation (185delAG) in the
BRCA1 gene in Ashkenazi Jewish women with a family history of breast
cancer led to other studies in Jewish individuals unselected for
family history. In a study of 858 samples taken from Ashkenazi Jewish
individuals seeking genetic testing for conditions unrelated to
cancer and 815 control individuals not selected for ethnic origin,
the 185delAG mutation was found in 8 of the 858, or nearly 1.0% of
Ashkenazim, and in none of the control samples.[18] The observed
carrier frequency of this specific mutation is several times higher
than the expected frequency of all BRCA1 mutations combined in the
general population.

It is thought that the 185delAG Ashkenazi Jewish mutation may have
arisen in an ancestral "founder" several generations ago,
dating back to approximately ad 1200.[19] (A founder mutation is an
altered gene seen with high frequency in a population originating
from a small ancestral group, one or more of the founders of which
was a carrier of the mutant gene.) Several mutations with founder
effects have been described in other ethnic groups, including
French-Canadians, Icelanders, and African-Americans.[20-22] In
Canada, a common origin of two BRCA1 mutations was also found in
breast and ovarian cancer families.[20] One of these mutations was
the 185delAG and the other was the 5382insC in codon 1756. This
latter mutation appears more frequently in Northern and Eastern
Europeans and may also be more common in Ashkenazi Jews.]

A single base-pair deletion in BRCA2 (6174delT) has also appeared
recurrently in Ashkenazi Jews, accounting for 8% of early-onset
breast cancer.[23] This BRCA2 mutation, together with the 185delAG
mutation in the BRCA1 gene, may account for up to 30% to 50% of all
early-onset breast or ovarian cancer cases in the Ashkenazim. The
three distinct recurrent mutations found in the Ashkenazi Jewish
population and their carrier rates are outlined in
Table 3

Studies in Other Groups--Recurrent mutations have been
described in other ethnic groups including, Icelanders, African-Americans,
and the Dutch. In studies in Icelandic families, a five base-pair
deletion (999del5) in BRCA2 was found in 16 of the 21 families
studied. Twelve of the mutation carriers were males with breast
cancer, accounting for 40% of all males diagnosed with breast cancer
in Iceland over the past 40 years.[21]

Based on an analysis of families with early-onset breast and ovarian
cancer, Gao et al reported that BRCA1 mutations may explain an
increased susceptibility to breast cancer in high-risk
African-American families. Two BRCA1 mutations (1832del5 and
5296del4) each were observed in two unrelated high-risk African
American families, as ascertained through young breast cancer cases
in an urban cancer risk clinic.[22]

In the Netherlands, a mutation (2804delAA) in BRCA1 accounted for 24%
of all mutations studied.[25] Other studies in this population
suggested that the BRCA1 and BRCA2 mutations have a significant role
in Dutch high-risk breast cancer families, with no evidence of any
additional highly penetrant breast cancer genes in these families.

Other Breast Cancer Susceptibility Genes

TP53 Gene--The presence of germ-line TP53 gene alterations has
been observed in families with the Li-Fraumeni syndrome. Families
with this syndrome have a variety of inherited cancers, including
breast and brain cancers, sarcoma, leukemia, and adrenocortical
cancers. The p53 protein plays a major role in the transcription
("reading") of DNA, cell growth and proliferation, and a
number of metabolic processes. Since p53 suppresses abnormal cell
proliferation, it may represent an important protective mechanism
against cancer. The p53 protein also appears to be involved in
programmed cell death, or apoptosis. It has been estimated that
approximately 1% cases of of inherited breast cancers are due to
mutations in the TP53 gene.

The PTEN gene on chromosome 10 has been shown to be mutated in
families with Cowden’s syndrome. Women from families with this
syndrome have a 30% to 50% lifetime risk of breast cancer.
Cowden’s syndrome is characterized by multiple hamartomas,
trichilemmomas, hyperkeratosis and mucocutaneous papillomatosis, and
an increased risk of thyroid and breast cancers. It has been
estimated that PTEN gene mutations may account for about 1% of cases
of inherited breast cancers.[26]

ATM Gene--Women who carry the ATM (ataxia telangiectasia
mutated) gene have been shown to be more sensitive to ionizing
radiation, which may lead to breast cancer. Ataxia-telangiectasia is
a rare autosomal-recessive disease characterized by chromosome
fragility, progressive cerebellar ataxia, ocular apraxia,
telangiectasias, and humoral and cellular immune deficiency.

It has been proposed that ATM heterozygotes have a predisposition to
cancer, with possibly a three- to fourfold increased risk in general
and a fivefold increased risk for breast cancer in women.[27] There
have been few studies documenting ATM mutations in high-risk
families. Thus, the proportion of breast cancer patients with ATM
mutations appears to be extremely low.


Loading comments...
Please Wait 20 seconds or click here to close